Ludwig and Jacob (2009).Pdf
Total Page:16
File Type:pdf, Size:1020Kb
Changing Poverty, Changing Policies Cancian, Maria, Danziger, Sheldon Published by Russell Sage Foundation Cancian, Maria. and Danziger, Sheldon. Changing Poverty, Changing Policies. New York: Russell Sage Foundation, 2009. Project MUSE. Web. 7 Jul. 2015. http://muse.jhu.edu/. For additional information about this book http://muse.jhu.edu/books/9781610445986 Access provided by University of Wisconsin @ Madison (16 Nov 2015 12:04 GMT) Chapter 10 Improving Educational Outcomes for Poor Children Brian A. Jacob and Jens Ludwig ne of the best ways to avoid being poor as an adult is to obtain a good edu- cation. As Katherine Magnuson and Elizabeth Votruba-Drzal’s chapter in Othis volume notes, people who have higher levels of academic achievement and more years of schooling earn more than those with lower levels of human cap- ital. This is not surprising in light of the belief by economists that schooling makes people more productive and that wages are related to productivity. Yet in modern America poor children face an elevated risk for a variety of adverse educational outcomes. In the 2007 National Assessment of Educational Progress (NAEP), only 16 percent of fourth-grade students eligible for free lunch scored at proficient levels in reading, compared with 44 percent of fourth graders whose family incomes were above the eligibility cutoff for free lunch; the disparity in math scores was even larger, 21 percent versus 53 percent (NCES 2007a; NCES 2007b). Equally large disparities in achievement test scores are observed between whites and minority racial or ethnic groups, with test score gaps that show up as early as three or four years of age (Jencks and Phillips 1998).1 In fact, the black- white test score gap among twelfth-graders may not be all that different in mag- nitude from the gap observed among young children when they first start school (Phillips, Crouse, and John 1998; Ludwig 2003). Understanding why children’s outcomes vary so dramatically along race and class lines in America is central to formulating effective education policy interven- tions. Disagreements about how to improve schooling outcomes for poor children stem in part from different beliefs about the problems that underlie the unsatis- factory outcomes in many of our nation’s public schools. Broadly speaking, critics tend to invoke, at least implicitly, one of the following reasons why children in high-poverty schools are not performing as well as we would like: 1. Schools serving poor and minority students have fewer resources than they need. In this case, a potential solution would be to provide more money to dis- advantaged schools. 266 / Improving Educational Outcomes for Poor Children 2. High-poverty schools lack the capacity to make substantial improvements in student learning, independent of financial resources. Under this perspective, the teachers and administrators in highly disadvantaged school districts are thought to lack the skills or knowledge necessary to improve the quality of instruction on their own. Potential solutions to this problem would involve helping schools improve the quality of their standard operating practices, for example, by helping to implement specific new instructional or organizational practices (that is, curriculum, instruction, school organization) or by increasing the instructional capacity of staff in these schools through professional devel- opment or more selective hiring. The success of this approach depends criti- cally on whether there exist particular instructional, organizational, human resources, or other practices that are known to be effective and on how easily these practices could be implemented across different contexts. 3. High-poverty schools do not have sufficient incentives and/or flexibility to improve instruction. High-poverty schools underperform because teachers and administrators are not working hard enough, they are not working toward the right goal, or they have good local knowledge about what would work best but are unable to implement these ideas because of centralized authority. Proponents of this perspective often claim that without clarifying the key objec- tives of schools and holding key actors accountable, additional spending is sim- ply squandered. Their solution is to enhance incentives and provide local actors with more flexibility through policies such as school choice or accountability. 4. Schools matter only so much. The real problem rests with the social context in which schools operate—namely, the family, neighborhood, and peer environ- ments that under this perspective make it difficult for low-income children to take advantage of educational opportunities. Adopting accountability or market- oriented reforms without changing social policy more broadly simply punishes educators for factors beyond their control and may drive the most able teachers toward schools that serve less disadvantaged students. For some reason, current education policy debates often seem to be argued as if these problems are mutually exclusive. In contrast, we believe that there is probably some truth to each of these major explanations—in other words, there is no single problem with our schools (and so potentially no single solution). Identifying the optimal policy response to the mix of problems that plague our public schools is complicated by the possibility that these problems might interact with each other. For example, it may be the case that certain curriculum reforms are effective only if they are accompanied by an increase in resources such as student support serv- ices, or by an increase in teacher quality generated by reforms to hiring and tenure policies. Social science theory and common sense are likely to carry us only so far in identifying the most effective—and most cost-effective—mix of education pol- icy changes. For almost every education intervention that some theory suggests might be effective, there is another plausible theory suggesting that the interven- tion will be ineffective or even harmful. Education policy also needs to be guided by rigorous evaluation evidence about what actually works in practice. / 267 Changing Poverty, Changing Policies Evaluation research in education has typically focused on easily measured and relatively short-term student outcomes, such as achievement test scores or school attainment. For this reason, most of the evidence we discuss here emphasizes these outcomes. In general, we expect these outcomes to be positively correlated with other important but harder-to-measure educational outcomes that society cares about, such as a student’s interest in school or classroom behavior. But in some cases interventions could have different types of impacts on easy-to-measure versus harder-to-measure outcomes, as might occur, for example, with current proposals to shift Head Start’s emphasis from comprehensive services to more academically oriented instruction. In these cases, understanding impacts on test scores or attain- ment may not be sufficient, but is still useful since these educational outcomes are associated with both a child’s chances of becoming poor as an adult and other life outcomes that society cares about. More generally, in principle there is no reason why education research could not track other outcomes, such as a student’s inter- est in school or a student’s classroom behavior. Research over the past four decades has unfortunately fostered the impression that when it comes to improving schools for poor children, “nothing works.” One of the first studies to contribute to this sense of pessimism was the landmark report by the sociologist James Coleman and his colleagues (1966). Drawing on a large, nationally representative sample, the so-called Coleman Report found that most of the variation in student test scores occurred within rather than across schools, that family background was the strongest predictor of academic achievement, and that most measurable school inputs, like student-teacher ratios, were only weakly correlated with student outcomes. Subsequent evaluation studies of different edu- cational interventions also tended to be disappointing and contributed to the pes- simism about the ability of schools to increase poor children’s life chances (Levin 1977; Glazer 1986; Jencks 1986). In contrast, our chapter offers a message of tempered optimism. In the past few decades, dramatic improvements in the technology of education policy evaluation have enhanced our ability to uncover moderately sized program impacts within the complex environment that determines schooling outcomes. The available evidence reveals a number of potentially promising ways to improve the learning outcomes of low-income children. This is not to say that everything works: many current and proposed education policies either enjoy no empirical support for their effective- ness or in some cases enjoy strong empirical evidence for their ineffectiveness. But a careful sifting of the empirical evidence identifies a selected set of interventions that seem to be promising. Our optimism is tempered by the recognition that the most successful educa- tional interventions reduce, but do not eliminate, racial and social class dispari- ties in educational outcomes. This is not a reason for either despair or inaction. The appropriate standard of success for policy interventions is that they generate net benefits, not miraculous benefits. Education policies that are capable of improv- ing poor children’s schooling outcomes by enough to justify the costs of these policies are